Defect Detection Research Articles

Phased Array Ultrasonic Inspection of Composite Fuselage Panel

Abstract The composite fuselage panel is a crucial component in large civil aircraft, necessitating higher detection rates and efficiency. Phased array ultrasonic testing can enhance the efficiency and defect detection rate; however, it is essential to consider the structural characteristics of the fuselage to ensure reliable identification of design defects. Based on the material properties, structural features, and defect detection requirements of composite fuselage panels, a test block reflecting their characteristics was designed and fabricated with various prefabricated defects. Ultrasonic phased array method was employed for a series of testing parameter optimization experiments and defect detectability analysis. The detected flaws were quantitatively analyzed and compared with those obtained using conventional ultrasonic automatic scanning techniques to validate the effectiveness of the developed ultrasonic phased array method. The feasibility of utilizing phased array ultrasonics for nondestructive evaluation of wide-body airliner composite fuselage panels is confirmed.

Ultrasonic evaluation of wire arc additively manufactured parts with heterogeneous microstructure

Abstract In Wire Arc Additive Manufacturing (WAAM) components manufactured by gas metal arc welding strategy, periodically alternating patterns of fine- and coarse-grained microstructures develop along the building direction, depending on the temperature evolution. Ultrasonic non-destructive evaluation relies on the scattering of waves from defects. In the material microstructure, ultrasound waves are also scattered at the grain boundaries as grain noise. The work presents a study on the influence of heterogeneous microstructure in WAAM on defect detectability during ultrasonic inspections. The study used a 2D finite element model to simulate ultrasound wave propagation in the material microstructure represented by a surface element from electron backscatter diffraction scans. Side-drilled holes were introduced in the material as defects. Phased array was used to perform full matrix capture and obtain ultrasound image through total focusing method, with dynamic focusing capability to inspect the heterogeneous microstructure. The study analysed the defect echoes and noises using the signal-to-noise ratio (SNR) to assess its defect detectability. While defects were well detected, the SNR values differ by 10% for defects at various positions along the building and transverse directions in the heterogeneous microstructure.

Effects of transducer’s Parameters on the Amplitude Response of Spirally Propagating Circumferential Lamb Wave and Their Application in Defect Detection

Abstract The commonly used traditional cylindroid guided wave modes are not sensitive to a crack with limited circumferential extent within a pipe. The circumferential guide wave will not be affected by the limited circumferential extent of a crack, which can be make up the limitation of the commonly used traditional cylindroid guided wave modes in crack detecting within a pipe. This article studied the spirally propagating circumferential Lamb wave within a pipe, which can determine the axial position of a crack only if the axial extent of the crack is large enough regardless the limited circumferential extent of the crack. To efficiently trigger the spirally propagating circumferential Lamb wave, the effect of the frequency, the circumferential coverage and the axial length of the transducer on the spirally propagating circumferential Lamb wave signals were investigated based on the theoretical model and experiments. Then, the comparison experiments presented the superiority of the proposed spirally propagating circumferential Lamb wave in detecting a crack with limited circumferential extent over the commonly used cylindroid guided wave mode. Finally, the angular interval of two cracks that in the same axial position can be determined through the angular interval of the triggering positions of the spirally propagating circumferential guided wave, which cannot be evaluated through the commonly used traditional cylindroid guided wave modes.

Machine learning-based decision-making approach for predicting defects detection: a case study

In today’s highly competitive global market, industries must produce faultless products to achieve profitability. Machine learning (ML) algorithms provide a possible method to improve quality standards by enabling the prediction of the outcome of quality control processes. This article presents a real case study based on ML algorithms suggested to develop a knowledge-based intelligent supervisory system to predict defect products in the fashion industry. Defect detection is formulated as a binary classification problem, and several ML algorithms have been compared to determine the most suitable one on the available data. The random forest (RF), LightGBM, and C5.0 algorithms exhibit comparable high-end performances on the pre-processed dataset made available by the company. Nevertheless, since the aim of the analysed industry is to reduce the rate of false negative observations (i.e., the proportion of defected-free products wrongly classified), the best method results is RF, as it minimizes this metric.

Search and recovery network for camouflaged object detection

Camouflaged object detection aims to accurately identify objects blending into the background. However, existing methods often struggle, especially with small object or multiple objects, due to their reliance on singular strategies. To address this, we introduce a novel Search and Recovery Network (SRNet) using a bionic approach and auxiliary features. SRNet comprises three key modules: the Region Search Module (RSM), Boundary Recovery Module (BRM), and Camouflaged Object Predictor (COP). The RSM mimics predator behavior to locate potential object regions, enhancing object location detection. The BRM refines texture features and recovers object boundaries. The COP fuse multilevel features to predict final segmentation maps. Experimental results on three benchmark datasets show SRNet's superiority over SOTA models, particularly with small and multiple objects. Notably, SRNet achieves performance improvements without significantly increasing model parameters. Moreover, the method exhibits promising performance in downstream tasks such as defect detection, polyp segmentation and military camouflage detection.

The importance of image quality in AI visual inspection systems

AbstractThe use of AI for visual inspection is currently making great strides in sectors like automotive, metals and machinery manufacturing. However, those investing in the technology not only expect a system that will be superior to manual inspection, but an error rate close to zero in defect detection. Let us look at the importance of image quality and why the variability in the production environment is the single greatest challenge that AI engineers must face.

Methods for Control and Detection of Fire-Hazardous Defects of Contacts and Contact Connections During Operation of Electrical Installations

Methods for Control and Detection of Fire-Hazardous Defects of Contacts and Contact Connections During Operation of Electrical Installations

Efficiency analysis of laser ultrasonic longitudinal wave detection based on the constraint effect

Abstract Laser ultrasound is a non-destructive inspection technique, but its application in internal defect detection and state characterization is limited by its weak body wave displacement induced by thermo-elastic mechanism. In this paper, the influence of the constraint layer on the directivity of laser induced ultrasonic force source and the excitation efficiency of longitudinal wave is analyzed by finite element modeling with transparent glass as the constraint layer. With the optimization of the parameters of the confinement layer, the longitudinal waves with energy comparable to that of the free surface ablation mechanism can be excited under the thermoelastic mechanism, which greatly improves the efficiency of the longitudinal wave of laser ultrasonic. Applying the above research results to the detection of internal defects, the imaging results of the internal defects are obvious and the location defects are accurate.

Simulation and experiment of eddy current detection of U-shaped defects in steel plates

Simulation and experiment of eddy current detection of U-shaped defects in steel plates

Ultrasonic testing method for impeller brazing defects with refraction wave perpendicular to brazing surface and angle amplitude compensation

Abstract There are blind areas in ultrasonic brazing defect detection, due to the non-parallel brazing surface and outer surface for many components including impeller. In this paper an oblique incident ultrasonic testing method with a refraction wave perpendicular to brazing surface is proposed. Meanwhile, an incident angle amplitude compensation is implemented to decrease the influence of the defect echo amplitude difference caused by oblique incidence and unify the sensitivity of different scanning positions. The compensation curve based on the theoretical curve of echo energy at different incident angles is optimized with iteration of actual measured defect echo amplitude. In order to prove the need for ultrasonic testing method with a refraction wave perpendicular to brazing surface, bottom surface echo test for conical samples is operated. In addition, with the method proposed in this paper, scanning image of impeller sample with the same detection sensitivity is obtained successfully. Testing results suggest that in relation to old method with incident wave perpendicular to the outer surface, the method proposed in this paper can solve the blind area problem effectively.

The numerical reconstruction methods for characterizing defects by full matrix data using an ultrasonic array

Abstract This study introduces ultrasonic imaging and inversion methods for the quantitative characterization of crack defects. The Total Focusing Method (TFM) is a high-precision ultrasonic imaging method, but it may produce length errors when used for large-angle defect detection. An SV transverse wave TFM method is introduced, significantly reducing length errors in a 60° crack to 8.31%. However, the ultrasonic image method is influenced by factors like array position, limiting its practicality. To overcome this limitation, the study proposes an inversion method using the subarray scattering coefficient matrix, which is more stable and less sensitive to array positioning. Comparative analysis with finite element simulations and experiments indicates optimal array positions associated with different methods. At these positions, both methods yield angular characterization errors for the 60° crack of less than 1%, validating the proposed techniques for non-destructive testing applications.

Multi-scale input reconstruction network and one-stage instance segmentation for enhancing heart defect prediction rate

<p>Artifacts and unpredictable fetal movements can hinder clear fetal heart imaging during ultrasound scans, complicating anatomical identification. This study presents a new medical imaging approach that combines one-stage instance segmentation with ultrasound (US) video enhancement for precise fetal heart defect detection. This innovation allows real-time identification and timely medical intervention. The study acquired 100 fetal heart US videos from an Indonesian Hospital featuring cardiac septal defects, generating 1,000 frames for training, validation, and testing. Utilizing a combination of the multi-scale input reconstruction network (MIRNet) for image enhancement and YOLOv8l-seg for real-time instance segmentation, the method achieved outstanding validation results, boasting a 99.50% mAP for bounding box prediction and 98.40% for mask prediction. It delivered a remarkable real-time processing speed of 68.4 frames per second. In application to new patients, the method yielded a 65.93% mAP for bounding box prediction and 57.66% for mask prediction. This proposed approach offers a promising solution to early fetal heart defect detection using ultrasound, holding substantial potential for enhancing healthcare outcomes.</p>

A New Feature Pyramid Network With Bidirectional Jump Connection Network for Small Defect Detection on Solar PV Wafer

A New Feature Pyramid Network With Bidirectional Jump Connection Network for Small Defect Detection on Solar PV Wafer

Road Surface Defect Detection—From Image-Based to Non-Image-Based: A Survey

Road Surface Defect Detection—From Image-Based to Non-Image-Based: A Survey

CC-De-YOLO: A Multiscale Object Detection Method for Wafer Surface Defect

Abstract Surface defect detection on wafers is crucial for quality control in semiconductor manufacturing. However, the complexity of defect spatial features, including mixed defect types, large scale differences, and overlapping, results in low detection accuracy. In this paper, we propose a CC-De-YOLO model, which is based on the YOLOv7 backbone network. Firstly, the coordinate attention is inserted into the main feature extraction network. Coordinate attention decomposes channel attention into two one-dimensional feature coding processes, which are aggregated along both horizontal and vertical spatial directions to enhance the network’s sensitivity to orientation and position. Then, the nearest neighbor interpolation in the upsampling part is replaced by the CAR-EVC module, which predicts the upsampling kernel from the previous feature map and integrates semantic information into the feature map. Two residual structures are used to capture long-range semantic dependencies and improve feature representation capability. Finally, an efficient decoupled detection head is used to separate classification and regression tasks for better defect classification. To evaluate our model’s performance, we established a wafer surface defect dataset containing six typical defect categories. The experimental results show that the CCDe-YOLO model achieves 91.0% mAP@0.5 and 46.2% mAP@0.5:0.95, with precision of 89.5% and recall of 83.2%. Compared with the original YOLOv7 model and other object detection models, CC-De-YOLO performs better. Therefore, our proposed method meets the accuracy requirements for wafer surface defect detection and has broad application prospects. The dataset containing surface defect data on wafers is currently publicly available on GitHub (https://github.com/ztao3243/Wafer-Datas.git).

Defect Recognition and Classification Using Ultrasonic Scattered Signals Based on Convolutional Neural Network

Abstract Carbon Fiber Reinforced Polymer (CFRP) is widely used in the aerospace, aircraft, and railway industries because of its remarkable advantages, such as its lightweight and high strength. Detecting and identifying defects in CFRP is crucial for ensuring its secure and efficient use. Ultrasonic non-destructive testing is one of the most practical and effective methods for defect detection and identification. The ultrasonic scattered signals of defects, which contain abundant defect feature information, are advantageous for defect recognition and classification. However, the anisotropy and inhomogeneity of CFRP, which lead to a low signal-to-noise ratio and poor image quality, pose challenges in flaw recognition. To address the issue, this paper proposes using Convolutional Neural Networks (CNN) methods for defect classification, which can adequately extract information from ultrasonic scattered signals directly. The scattering signals from various types of defects in CFRP, including the different sizes of holes and folds, were selected as inputs for CNN models. Different network models combining Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) with 1D-CNN were employed to classify and identify defect signals. To generate sufficient input sample signals, the data sets were further enlarged by incorporating noise into the original signals. The experimental results showed that using the 1D-LSTM-CNN and 1D-GRU-CNN network models can reduce computational resources while ensuring the accuracy of defect recognition.

Sparse-TFM imaging in frequency domain based on amplitude-squared parameter

Abstract The total focusing method (TFM) implemented in time domain has high complexity, limiting its large-scale engineering applications. In contrast, the frequency-domain TFM based on phase shift migration (PSM) is helpful to improve imaging efficiency. On this basis, partial data can be abandoned for imaging to save computing costs and shorten processing time without reducing imaging quality. In this paper, an adaptive sparse-TFM in the frequency domain is proposed by selecting desired signals from full matrix capture (FMC) data. The amplitude-squared parameter for all the received signals corresponding to each transmitting element is taken as the relevant weight, and sparse arrays are constructed adaptively according to the descending order of numerical values. Simulated and experimental results indicate that the proposed method has stable imaging quality under different sparsity ratios, rapidly improving imaging efficiency with good resolution for defect detection.

Application of optical fiber distributed acoustic sensing system in ultrasonic detection

Abstract The optical fiber distributed acoustic sensing system can continuously detect acoustic signals along the optical fiber, with the characteristics of wide response frequency band, large capacity, anti-electromagnetic interference and so on. An ultra-sensitive distributed optical fiber acoustic sensing system (uDAS) was produced with a high response frequency up to 100 kHz and thousands of sensing points. The arrayed ultrasonic detection can be realized by the distributed characteristics of the system. For verifying its ultrasonic detection ability, it is applied to ultrasonic nondestructive testing of structural defects. 20 optical fiber ultrasonic detection units were constructed. Each one is a fiber ring with a diameter of about 1.5 cm and a length of 2 meters. Two cubic cement structures with a side length of 30 cm were manufactured for experiment, one of which had an artificial internal defect. The optical fiber array was attached to the surface of the cement structure, and the ultrasonic signal with a frequency of 40 kHz was excited at a single point on the opposite side. The ultrasonic propagation speed changes on account of the defects such as holes and loose texture in the propagation path. The defects in the structure are identified and located by analyzing the difference of arrival time of waves. It is expected that such sensing system could found important applications in structure defect detection.

In-situ defect detection in laser-directed energy deposition with machine learning and multi-sensor fusion

In-situ defect detection in laser-directed energy deposition with machine learning and multi-sensor fusion

High-noise solar panel defect identification method based on the improved EfficientNet-V2

As a crucial element in photovoltaic power generation systems, the condition of solar panels significantly impacts the efficiency of power generation. The ability to accurately and promptly detect defects in solar panels is essential for enhancing system performance. This study introduces a novel model for identifying defects in photovoltaic modules, leveraging an enhanced version of EfficientNet-V2. This model aims to address challenges in identifying defects in infrared images of solar panels under conditions of high-noise and low-model efficiency. To address the challenges of high image noise and blur, this article initially presents a methodology that combines the Db4 wavelet transform with a blind deconvolution algorithm for comprehensive preprocessing of the original image. Furthermore, this study optimizes the model's feature representation capabilities by implementing key transformations within the EfficientNet-V2 network framework. Notably, we replaced the traditional SE block with the more efficient channel attention (ECA) mechanism module. Due to its lightweight structure and effective performance, ECA substantially improves the model's capacity to extract complex and abstract image features, while also accelerating the training process's convergence speed and enhancing overall computational efficiency. At the classifier level, this paper innovatively integrates the XGBoost ensemble learning algorithm into the model, substituting the conventional softmax classifier used in traditional convolutional neural network (CNN). With its superior generalization capabilities, robust nonlinear modeling skills, and efficient computational characteristics, XGBoost can more accurately detect minute defects in solar panels based on the deep features produced by EfficientNet-V2, thereby significantly improving the accuracy and robustness of defect detection. Simulation results demonstrate that the proposed model structure outperforms traditional CNN models in terms of accuracy and stability, underscoring the efficacy of the enhanced EfficientNet-V2 model in detecting solar panel defects under high-noise conditions.